{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:FRVIVM4OFU3YYEBD425EMZMTUF","short_pith_number":"pith:FRVIVM4O","schema_version":"1.0","canonical_sha256":"2c6a8ab38e2d378c1023e6ba466593a14afc956feb33eb02cae466bde431a789","source":{"kind":"arxiv","id":"2505.08910","version":2},"attestation_state":"computed","paper":{"title":"Behind Maya: Building a Multilingual Vision Language Model","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Abhipsha Das, Alham Fikri Aji, Anthony Susevski, Ashvanth.S, Bala Krishna S Vegesna, Chen Liu, Drishti Sharma, Genta Indra Winata, Isha Chaturvedi, Karthik Reddy Kanjula, Nahid Alam, Roshan Santhosh, Ryan Sze-Yin Chan, Shayekh Bin Islam, S M Iftekhar Uddin, Snegha A, Snehanshu Mukherjee, Surya Guthikonda, Timothy Chung","submitted_at":"2025-05-13T19:01:12Z","abstract_excerpt":"In recent times, we have seen a rapid development of large Vision-Language Models (VLMs). They have shown impressive results on academic benchmarks, primarily in widely spoken languages but lack performance on low-resource languages and varied cultural contexts. To address these limitations, we introduce Maya, an open-source Multilingual VLM. Our contributions are: 1) a multilingual image-text pretraining dataset in eight languages, based on the LLaVA pretraining dataset; and 2) a multilingual image-text model supporting these languages, enhancing cultural and linguistic comprehension in visio"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2505.08910","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-13T19:01:12Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"4f564110db8b00fdde36f0d6079aeb52810401e6780f40c2b3c9205a48ce6376","abstract_canon_sha256":"72dc47a3931b78382567083baa135049c4159f698b446657da475156a7e7b3c7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:03:25.049061Z","signature_b64":"uuakjuRf8xieEujvnxGA/0DOQPRmXyyKK/dMDkx4BOhEl/jyTakE8LjONwckyHLvNHG9EJJOMD2ohl8kP3D6DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2c6a8ab38e2d378c1023e6ba466593a14afc956feb33eb02cae466bde431a789","last_reissued_at":"2026-07-05T11:03:25.048488Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:03:25.048488Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Behind Maya: Building a Multilingual Vision Language Model","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Abhipsha Das, Alham Fikri Aji, Anthony Susevski, Ashvanth.S, Bala Krishna S Vegesna, Chen Liu, Drishti Sharma, Genta Indra Winata, Isha Chaturvedi, Karthik Reddy Kanjula, Nahid Alam, Roshan Santhosh, Ryan Sze-Yin Chan, Shayekh Bin Islam, S M Iftekhar Uddin, Snegha A, Snehanshu Mukherjee, Surya Guthikonda, Timothy Chung","submitted_at":"2025-05-13T19:01:12Z","abstract_excerpt":"In recent times, we have seen a rapid development of large Vision-Language Models (VLMs). They have shown impressive results on academic benchmarks, primarily in widely spoken languages but lack performance on low-resource languages and varied cultural contexts. To address these limitations, we introduce Maya, an open-source Multilingual VLM. Our contributions are: 1) a multilingual image-text pretraining dataset in eight languages, based on the LLaVA pretraining dataset; and 2) a multilingual image-text model supporting these languages, enhancing cultural and linguistic comprehension in visio"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.08910","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2505.08910/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2505.08910","created_at":"2026-07-05T11:03:25.048559+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.08910v2","created_at":"2026-07-05T11:03:25.048559+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.08910","created_at":"2026-07-05T11:03:25.048559+00:00"},{"alias_kind":"pith_short_12","alias_value":"FRVIVM4OFU3Y","created_at":"2026-07-05T11:03:25.048559+00:00"},{"alias_kind":"pith_short_16","alias_value":"FRVIVM4OFU3YYEBD","created_at":"2026-07-05T11:03:25.048559+00:00"},{"alias_kind":"pith_short_8","alias_value":"FRVIVM4O","created_at":"2026-07-05T11:03:25.048559+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.26601","citing_title":"FTibSuite: A Comprehensive Resource Suite for Tibetan Vision-Language Modeling","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FRVIVM4OFU3YYEBD425EMZMTUF","json":"https://pith.science/pith/FRVIVM4OFU3YYEBD425EMZMTUF.json","graph_json":"https://pith.science/api/pith-number/FRVIVM4OFU3YYEBD425EMZMTUF/graph.json","events_json":"https://pith.science/api/pith-number/FRVIVM4OFU3YYEBD425EMZMTUF/events.json","paper":"https://pith.science/paper/FRVIVM4O"},"agent_actions":{"view_html":"https://pith.science/pith/FRVIVM4OFU3YYEBD425EMZMTUF","download_json":"https://pith.science/pith/FRVIVM4OFU3YYEBD425EMZMTUF.json","view_paper":"https://pith.science/paper/FRVIVM4O","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.08910&json=true","fetch_graph":"https://pith.science/api/pith-number/FRVIVM4OFU3YYEBD425EMZMTUF/graph.json","fetch_events":"https://pith.science/api/pith-number/FRVIVM4OFU3YYEBD425EMZMTUF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FRVIVM4OFU3YYEBD425EMZMTUF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FRVIVM4OFU3YYEBD425EMZMTUF/action/storage_attestation","attest_author":"https://pith.science/pith/FRVIVM4OFU3YYEBD425EMZMTUF/action/author_attestation","sign_citation":"https://pith.science/pith/FRVIVM4OFU3YYEBD425EMZMTUF/action/citation_signature","submit_replication":"https://pith.science/pith/FRVIVM4OFU3YYEBD425EMZMTUF/action/replication_record"}},"created_at":"2026-07-05T11:03:25.048559+00:00","updated_at":"2026-07-05T11:03:25.048559+00:00"}